For expository convenience, the present disclosure is illustrated with reference to audio content. However, it should be recognized that the principles described below are applicable in any media context, including still imagery, video, product packaging, etc.
In the cited patent applications, the present assignee disclosed a variety of technologies by which audio content can be associated with corresponding internet resources. In some such approaches, the audio content is steganographically encoded (e.g., by digital watermarking) to convey an identifier. When a computer encounters such an encoded audio object, it discerns the encoded identifier, forwards the identifier to a remote database (a “Registry database”), and receives in response—from a database record indexed by the identifier—the address of one or more internet resources related to that audio (e.g., fan sites, concert schedules, e-commerce opportunities, etc.) The computer can then link to such a resource and present same to a user, e.g., using an internet browser program. Such an arrangement is shown in FIG. 1.
There are many variations on this model. For example, instead of steganographically encoding the identifier in the content, the identifier can be added into header or other data with which the content is conventionally packaged.
The identifier can be assigned to the content. Or the identifier can be derived, in some manner, from the content.
In the former, assigned identifier case, an entity such as a music publisher (e.g., Sony) or a music distributor (e.g., emusic.com), selects a number for encoding into the content. The number may be selected from a limited range of numbers (e.g., a range of numbers allocated to that publisher by the proprietor of the Registry database), but the number itself is not inherently related to the content with which it is associated.
In the latter case, the identifier is derived from the content, or from other information associated with the content.
One way to derive an identifier is to employ selected bits of the content, itself, as the identifier. For example, in MP3 audio, where the signal is encoded into frames, the Nth bit of the first 128 frames of a musical work can be assembled together into a 128 bit identifier. Or data present in MP3 headers can be used. In another approach, some or all of the content data is processed by a hashing algorithm to yield a 128 bit identifier corresponding to that content. In both of these cases, the identifier is implicit in the audio itself. That is, no data needs to be added (e.g., in a header, or by steganographic encoding).
When deriving the identifier from associated information, one can use the table of contents (TOC) of the CD or file allocation table of the DVD. The ID can be embedded within the MP3 file, if it is being ripped from the CD or DVD at the time or ID creation. The method of embedding can be embedded via header, footer or frame bits, or via a watermark.
Some techniques for deriving an identifier may rely on external resources. For example, when “ripping” a song from a commercial music CD into MP3 form, many ripper software programs refer to an on-line disc recognition database, found at www.cddb.com, to obtain the title and length of the song. This “table of contents” information can be used to form the identifier, e.g., by selecting predetermined bits, hashing, etc. In this case, the identifier must generally be added to the audio (i.e., it is explicit, as was the case of the assigned identifiers). Such an arrangement is shown in FIG. 2.
The artisan will recognize that there are an essentially infinite number of algorithms by which such derived identifiers can be generated. (It will be noted that derived identifiers may not be unique. That is, two unrelated audio files may—coincidentally—correspond to the same identifier. But by making the identifier sufficiently long (e.g., 128 bits), such occurrences can be made arbitrarily unlikely.)
When an identifier is assigned to content, the entity doing the assigning (e.g., a record label such as Sony, or a music distributor such as Emusic) can ensure that the Registry database has a record corresponding to that identifier. The database record contains, e.g., one or more URL(s) leading to information relating to the audio content.
A different situation arises when the identifier is derived from content. No master authority ensures that the Registry database has a record corresponding to that identifier. Thus, if a college student rips music from a privately-produced CD into an MP3 file, the identifier derived from that music may not point to an active database record in the Registry database. This can also occur with CDs from major or minor labels that don't register all their identifiers. For example, the Kinks' CDs may not be registered by the label owning rights to their albums because of their fall in popularity. However, a general consumer/business person could register the CD's identifier to sell Kinks' paraphernalia. The consumer/business person does not need to make the same amount of revenue as the record label to make the registration process and maintenance worth his/her time.
If the MP3 file so-produced becomes popular, and is widely spread (e.g., through means such as Napster, Gnutella, etc.) a large potential audience may develop for internet resources related to that MP3. The issue then arises: who manages the Registry database address represented by the corresponding identifier?
The present invention addresses this and related issues.
In accordance with one aspect of the present invention, a user who encounters an unused Registry database record is given an option to manage it, or to participate financially in its exploitation.
For example, in one embodiment, when the Registry database first receives a query corresponding to an un-used identifier, the person initiating the query is given an opportunity to lease that identifier for a predetermined period, such as two months. Upon payment of a nominal fee (e.g., $10), the user can specify a URL that will be stored in the Registry database in association with that identifier, and to which subsequent users will be directed.
In accordance with another aspect, when a user first queries an un-used identifier in the Registry database, an auction commences, with a nominal opening bid (e.g., $10). The auction continues for a short period, such as a week or a month, allowing other persons who encounter such music early in its distribution life to have a chance at gaining the leasehold rights. At the end of the auction, the winner is granted a lease to that identifier for a predetermined period and can specify the URL with which that identifier is associated.
At the end of the predetermined period, the identifier can be leased for a subsequent term—either for a fixed fee (e.g., a multiple of the fee earlier charged), or through an auction.
It will be recognized that this arrangement has certain similarities to the present system for internet domain name registration. A user can query a whois database maintained by Network Solutions and the like to determine whether a domain name is assigned. If it is not, Network Solutions will offer to assign the domain name for a term of years in exchange for a payment. But the present invention serves different needs and is otherwise different in certain details.
The foregoing and other features and advantages of the present invention will be more readily apparent from the detailed description, which proceeds with reference to the accompanying drawings.